The Agent Skills Directory

[SAFE]: The skill provides standard, well-documented templates for Discord bot development. It explicitly warns against anti-patterns like hardcoding secrets or requesting excessive permissions.
[DYNAMIC_EXECUTION]: The skill uses dynamic loading patterns for commands and extensions (e.g., require(filePath) in JavaScript and bot.load_extension in Python). These are standard architectural patterns for modular Discord bots and target local directories (./commands, ./cogs), presenting no significant security risk in this context.
[INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface by ingesting user data through Discord interactions (slash command options and messages). However, the provided examples use this data for simple responses without passing it to dangerous sinks like shell execution or file system operations.
Ingestion points: Discord interaction objects and application command options in src/commands/ and main.py.
Boundary markers: None explicitly implemented; the code uses standard string interpolation.
Capability inventory: No subprocess calls, file writes, or sensitive network operations are present in the provided scripts.
Sanitization: Relies on standard Discord API formatting and library-level escaping.

discord-bot-architect